In an exciting development for developers worldwide, February 2025 saw the unveiling of Visual Studio Code 1.98, featuring significant advancements in the GitHub Copilot AI coding assistant. This release marks a notable expansion of the capabilities of GitHub Copilot with several new features now available in preview. Visual Studio Code 1.98, announced on March 5, 2025, can be accessed on Linux, Windows, and Mac through the official website, and the GitHub Copilot extension for VS Code is available in the Visual Studio Marketplace.
Enhanced Functionality of GitHub Copilot
Copilot Edits Support for Notebook Files
A major highlight in this release is the enhanced functionality of Copilot, especially through the introduction of Copilot Edits support for notebook files. This feature provides developers with the ability to leverage Copilot’s assistance for editing notebook files, just as they would with traditional code files. Currently available only to users of VS Code Insiders with the pre-release version of GitHub Copilot Chat, Copilot Edits seamlessly integrates with both code and notebook editors. This enhancement is expected to significantly streamline the development process for data scientists and researchers who frequently work with both types of files, simplifying their workflow by providing a consistent editing experience across different file formats.
Moreover, the new update introduces a fixed viewport during changes, eliminating the need for cumbersome scrolling in large notebooks. This aspect greatly improves the user experience, making it effortless to implement changes without losing your place in the file. The renaming of edit review actions to “Keep” and “Undo” adds clarity and straightforwardness to the process, ensuring that users can easily follow and manage their edits. These refinements enhance the intuitive nature of Copilot, offering an even more user-friendly tool for both novice and seasoned developers.
Next Edit Suggestions and Advanced Code Base Searches
Another significant enhancement in this version is the introduction of a collapsed mode in Next Edit Suggestions (NES), which refines Copilot’s ability to predict subsequent edits. This feature allows developers to view suggested edits in a minimized format, facilitating a cleaner and more organized workspace. Copilot now makes it even easier to navigate large codebases by providing advanced search functionalities. For instance, developers can include #codebase in their chat queries to efficiently locate relevant code segments. This feature can be a game changer for large projects, significantly reducing the time spent searching for specific parts of the code.
Additionally, the new Copilot vision, currently in preview mode, allows developers to attach and interact with images within chat prompts. An example of this is the ability to highlight errors by attaching screenshots from VS Code, making it possible for Copilot to assist in identifying and rectifying these issues. This visual aspect introduces a new dimension to AI-assisted coding, helping developers quickly pinpoint and address problems within their projects. By integrating such comprehensive support, Copilot forges an indispensable space in the toolkit of modern developers.
User Experience Improvements
Experimental Agent Mode for Copilot Edits
Also notable is the introduction of the experimental agent mode for Copilot Edits, which was rolled out last month and continues to gain traction. This mode is designed to automate a range of tasks, such as searching a developer’s workspace, editing files, checking for errors, and executing terminal commands. With the latest update, several useful UX improvements are introduced, showing terminal commands inline, which makes it easier for developers to track and edit suggested commands on the fly. By having these commands visually accessible within the editor, developers can efficiently perform iterations and adjustments during their coding sessions, further streamlining the development process.
Terminal IntelliSense is another advancement included in this update that deserves mention. Currently in preview, Terminal IntelliSense enhances completion specifications, command-line parsing, along with folder and file completions across various shell environments like bash, zsh, fish, and PowerShell. This feature will be vital in optimizing how developers interact with the terminal, boosting productivity through smarter auto-completions and command hints. As developers increasingly rely on terminals for various tasks, the improved IntelliSense will contribute significantly to a more productive coding environment.
Additional Functionalities and Terminal Enhancements
Apart from the enhancements related directly to Copilot, Visual Studio Code 1.98 introduces several other noteworthy improvements, emphasizing broader utility and usability for developers. One such improvement is the enhancement of the editor’s peek view, which now fully supports drag-and-drop entries. Developers can now drag entries from their tree into separate editors, allowing for more flexible and dynamic manipulation of code elements. This functionality caters to a more intuitive and efficient editing process, helping developers manage their projects with greater ease.
In line with these user-friendly updates, the update brings forth a custom title bar set as the default on Linux. Additionally, an experimental commit hook has been introduced, which prompts users about unresolved diagnostics for changed files during commits. This feature aims to minimize errors and ensure that code integrity is maintained before finalizing changes. Another significant enhancement is made for TypeScript files, which now benefit from improved context for inline completions and /fix commands in Insider releases, activated by a specific setting in the application. These tailored improvements collectively enhance the coding experience, providing tools that help developers maintain cleaner, more efficient codebases.
Reflection on the Progress of VS Code
A Progressive Step from Previous Versions
Visual Studio Code 1.98 represents a substantial leap forward from its predecessor, VS Code 1.97, with a pronounced focus on enhancing GitHub Copilot accommodations and overall user experience. This release underscores Microsoft’s dedication to continuously refining developer tools, ensuring that the software remains cutting-edge and aligned with the evolving needs of the developer community. By integrating AI-driven assistance more deeply and introducing user-centric features, VS Code 1.98 sets a high bar for future releases.
The thoughtful integration of AI technology, as demonstrated by the enhanced GitHub Copilot features, indicates a strong trend toward AI-assisted coding in the industry. Copilot’s ability to predict code edits, conduct advanced searches, and interact with visual prompts exemplifies how AI can significantly bolster coding efficiency and accuracy. These advancements not only accelerate development processes but also reduce potential for errors, making it an invaluable asset for developers working on complex projects.
Enriched Developer Experience
In a thrilling update for developers around the globe, February 2025 showcased the launch of Visual Studio Code 1.98, which introduces substantial enhancements to the GitHub Copilot AI coding assistant. This release signifies a major leap in the functionality of GitHub Copilot, boasting several new features that are now available in preview mode. Officially announced on March 5, 2025, Visual Studio Code 1.98 is accessible across multiple platforms, including Linux, Windows, and Mac. Developers can download it from the official website, ensuring broad compatibility. Furthermore, the GitHub Copilot extension for Visual Studio Code is readily available for download in the Visual Studio Marketplace. This update is poised to revolutionize the way developers write and debug code, offering advanced AI support that aims to streamline workflow and enhance productivity. It reflects the ongoing commitment to pushing the boundaries of what coding assistants can achieve, making it an essential tool for developers looking to harness the full potential of AI in their coding endeavors.